Clustering techniques for data analysis and data completion of monitored structural responses of an offshore floating structure

Abstract

Offshore Floating Structures (OFSs) present a major category of offshore structures that are often subjected to severe environmental conditions and harsh critical loading scenarios. The state of an OFS during its life-cycle must remain in the domain specified in the design, although this can be altered by normal aging due to usage, the action of the environment and accidental events. In recent years, the field of Structural Health Monitoring (SHM) has been growing at a fast rate, especially in different applications within the offshore structures' field (e.g. platforms and systems in oil and gas technology, risers, and offshore wind technology). Based on the monitored data of the SHM a diagnosis and most importantly a prognosis of the health status of the OFS can be assessed. Usually, measured data in long time span of different structural response quantities are used for the aforementioned assessment with, in some cases, unmeasured data. This paper deals with two objectives for the case of monitored structural response data of an OFS: (i) the implementation of clustering techniques for analysis of the structural response data and (b) the completion of missing structural response data based on appropriate clustering techniques.